Using AI to provide a more powerful search experience can be difficult.
Lucidworks, a vendor of AI-powered search applications, is trying to make the process simpler and provide enterprises with a connected experience with a new SaaS platform called Springboard.
Thursday's release includes Springboard's first generally available application -- Connected Search.
Connected Search provides enterprises with a search and insight engine that includes push-button AI and guided and optimized workflows.
Enterprises that use Connected Search use the AI and machine learning capabilities of Springboard. Lucidworks said the push-button AI provides instant intelligence and improves relevancy by applying user insights and signals.
Lucidworks plans to release other applications for the Springboard platform throughout the year, including Connected Service in the third quarter and Connected Commerce in the fourth quarter.
The Connected Search application is listed to start at $600 per one million requests and 100,000 documents per month for early access customers.
The goal of the platform is to enable enterprises understand their customers' intent, said CEO Will Hayes.
In this Q&A, Hayes discussed what those connected experiences are and how Lucidworks' new search platform differs from the competition's products.
How did this search platform come about?
Will Hayes
Will Hayes: For the last eight years, we've been in business with what we call the fusion server. This is a search engine that's AI-powered with a primary focus around personalization and discovery. We take all the user signals as you're shopping, as you're browsing, adding things to your cart, contacting user support, we build a whole catalog. We run machine learning to use those signals to better target personalization and rank things for you. This is used by some of the larger retailers, some of the largest companies in grocery, home improvement, auto improvement, apparel and exercise gear.
We've had a lot of requests to take those capabilities -- that AI-powered research and personalization and recommendation -- and apply those to additional use cases such as customer service and e-commerce.
This is a search engine that's AI-powered with a primary focus around personalization and discovery.
Will HayesCEO, Lucidworks
To make that quicker and more impactful for our customers, we're launching a SaaS platform. This is a multi-tenanted SaaS platform that's application-specific.
What problem are you addressing with this search platform?
Hayes: The biggest differentiation is back to the way that we leverage what we call those user signals.
Unlike a lot of the competition that's focused on a particular sort of search use case, we believe that by gathering all those signals, running machine learning on all those signals, we can start to inform multiple channels.
One of the biggest differentiators and advantages that enterprises get through connected search is they're starting to collect all this user interaction data; this is what we call first-party data.
[Connected Search] is the site search application. This will allow customers to cater to your browsing experience and allow them to change results.
What makes this search platform different from your competitors?
Hayes: There's three kinds of key things. One of them is the way that we've applied AI within the technology. We have several different machine learning models that we use out of the box, for visualization and for recommendations. We have some that are trained on very domain-specific types of things. If you're in oil and gas or financial services, our AI can help better enrich the data, understand the data, answer user questions.
So as people are coming in and ask a question, you know, being able to surface and answer those questions, so we rely on something called semantic vectors.
Then if you look within our platform itself, we kind of have two key paradigms that we depend on to roll these use cases out.
The first one is our data services. Again, I talked a little bit about being specifically trained for specific domains. We're going into oil and gas or apparel, or home improvement. Those all have their own sets of languages in glossaries that they use that we can build data enrichment and understanding around. Then with our workflows, we're allowed to take all this magic, if you will, and put that into a process or a flow that, for instance, can help support a customer service agent.
Here's a simple example of a workflow. I'm a help desk agent; I get a phone call. I type in [the customer's name]. That workflow is going to know to pull up all products and services related to [the customer], pull up any open issues that are related to those products and services, pull up a resolution if we have a common issue that's occurring, and surface all that information in the moment to provide that workflow.
From the customer agent's [perspective], I just go on the screen, and I go, 'Oh, OK. I see the things that you've subscribed to. I see issues that you might be having, I see where you've been on the website, let me try to better service you.'
Bringing those things together -- the semantic vectors and AI, our data models that we have built around various domains and these workflows -- we have a unique way to go to market and deploy different use cases, using those same capabilities.
Hayes: There's two things about speed. One of them is sub-second indexing.
So, you give us a URL, we come in, we start calling that incredibly fast. We also update fast.
In most cases, with most solutions, there can be a delay of up to 12 to 24 hours before you see those updates show up in the index. Imagine if you're like a grocery provider, and inventories are changing constantly and you're needing to keep those indexes up to date. We can do that in real time, and we can do that at a very large scale.
Editor's note: This interview was edited for clarity and conciseness.
Will Hayes
